Enter the visitors and conversions for each variant to see the confidence level and whether the difference is statistically significant.
How it works
Enter the visitors and conversions for variant A (the control) and variant B (the challenger). The calculator runs a two-proportion z-test, the standard test for comparing two conversion rates, and reports the confidence level, the relative uplift, and a plain-English verdict.
It computes each rate, a pooled standard error, the z-score, and a two-tailed p-value. Confidence = (1 - p) x 100. A result is flagged significant at 95% confidence (p < 0.05).
Advanced options let you choose the confidence level (90, 95, or 99%) and a one or two-tailed test, and reveal the exact p-value plus the sample size you need per variant.
Everything runs in your browser. Your numbers are never sent to a server, and there is no signup or limit. Your last entries are remembered locally so the calculator is ready next time.
What you will see
- Confidence
- How sure you can be that the difference is real and not random noise. 95%+ is the usual bar.
- Result
- A verdict: significant at 95%, approaching significance, or not significant yet.
- Relative uplift
- How much better (or worse) variant B converts versus A, as a percentage of A.
Frequently asked questions
What does statistical significance mean in an A/B test?
It means the difference between your two variants is unlikely to be down to random chance. At 95% confidence there is only about a 5% probability you would see a gap this large if the variants truly performed the same.
How is significance calculated here?
With a two-proportion z-test. The tool compares the two conversion rates using a pooled standard error, converts the gap into a z-score, and derives a two-tailed p-value. Confidence is 1 minus that p-value.
What confidence level should I aim for?
95% is the common standard, and this calculator flags significance there. Some teams require 99% for high-stakes changes. Below 90%, treat the result as inconclusive and keep collecting data.
My result is not significant. What should I do?
Usually, keep the test running to gather more visitors, since larger samples detect smaller real differences. If the rates are nearly identical after a large sample, the change likely has little effect and you can move on.
Does this account for test duration or multiple variants?
No. It is a single two-variant (A vs B) significance check at one point in time. Run tests for full business cycles (typically one to two weeks minimum) to avoid day-of-week bias, and be cautious comparing many variants at once.
Is this A/B test calculator free?
Yes, free with no signup. It runs entirely in your browser, so your test data never leaves your device.